SIXTH REPORT OF THE DIRECTOR I909 165 



The very marked relation of these caves to the present water 

 levels of Lake Champlain strongly suggests this lake as their cause. 

 On the other hand, the vertical depth to which they descend below 

 low water, their increasing number at the lower levels, their great 

 width and horizontal depth, their appearance on fresh glaciated sur- 

 faces and their apparent modification by glacial action, all strongly 

 suggest a preglacial origin for them and leave them as rock in- 

 scriptions commemorative of lake Valcour. 1 



During the past season these caves have been studied with the 

 hope of finding evidence which shall serve in determining their age 

 and the manner of their origin. It is the purpose of this paper to 

 present the evidence found. The first of this evidence deals with a 

 peculiar feature of erosion found on all cave surfaces which have 

 been acted upon by the present lake waters. It will be necessary to 

 treat this feature somewhat in detail and this we propose to do 

 under the subtitle which follows. 



Dentpits. The beds of the uppermost Chazy are in part cov- 

 ered where they crop out at the northeast corner of the island, but 

 under the water of the lake and some distance out from the shore 

 line, wave action has swept them comparatively clean. During the 

 period of phenomenally low water in 1908 a series of samples of 

 these beds, which run from just below the sandstone capping of the 

 Chazy to the beds of the Black River epoch, were secured. One 

 bed of very pure dolomitic limestone of Lowville age ( ?) was 

 deeply cut on its upper surface by confluent cupholes 2 whose edges 

 were so sharp that it was an exceedingly painful matter to stand on 

 them barefooted even though the water was more than a meter 

 deep. This bed had been undercut, and a large block had fallen 

 from its edge and now lay in a number of smaller pieces. These 

 pieces however still remained in their proper relations to the bed 

 and to each other and none of them had been turned over by wave 

 action. On raising a piece of this bed weighing a little over 46 



The sink hole shown in plate 6 is in a meadow on the lower terrace 

 of the north farm and seems to be a still more recent opening. This 

 hole had been partially filled up with some old cedar stumps, and three 

 upright poles served to indicate its position to any one mowing the 

 field. These stumps and one of the poles were temporarily removed in 

 order to obtain a view of the rock edges in the opening. This opening 

 was not sufficiently illuminated to allow its interior to show any detail 

 in the photograph. 



1 See Some Items Concerning a New and an Old Coast Line of Lake 

 Champlain. N. Y. State Mus. Bui. 133, p. 159-63. 



2 loc. cit. p. 160-61, pi. 1-5 inc. 



